Collaborative filtering recommendation algorithm based on user ratings and item attributes

A collaborative filtering recommendation algorithm is proposed in recommender systems based on user ratings and item attributes to relieve the problem of data sparsity, which introduces the information of item attributes and considers the similarity of item attributes preference of users as well as the similarity of user ratings. It improves the accuracy of recommendation under the influence of data sparsity compared to the traditional collaborative filtering (CF) algorithm considering user ratings only. The experimental results of the proposed algorithm based on MovieLens data set are compared with the traditional CF algorithm and it is found that the mean absolute error (MAE) reduces by 0.08. The proposal improves the recommendation accuracy in recommendation of the commodity and social network and gives more satisfactory recommendation to users.